EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: Fixin Stuff on May 29, 2020, 08:11:21 am
-
UPDATE: Title updated 28th June 2020 - skip to page 3 to get to current issue. Thanks! :D
Hi Peeps! Welcome to my first EEVBlog post! :clap:
Been fixin' stuff all alone for years... but then in my quest for answers, I stumbled upon 'the promised land' that is EEVBlog!
I'm an electrician and a tinkerer. Got some basic electronics knowledge but am new to troubleshooting SMPS. This is the most ambitious repair I've undertaken so far. Doing it more for the fun/learning/lolz than the result. TBH, I don't know what I'd do with this TV if I get it working :-DD
While in lockdown a month or so ago and stricken with boredom, I discovered an LG47LE7500 47inch LCD LED TV (https://www.lg.com/au/tvs/lg-47LE7500-lcd-tv (https://www.lg.com/au/tvs/lg-47LE7500-lcd-tv)) in the trash room of my apartment building. After seeing it sitting there for a week and noticing the panel seemed to be in good condition, I managed to convince my girlfriend to allow me to drag it's heavy a$$ (the TV, not my GF) up to our apartment.
I plugged the TV in anddd...nothing. No noises, no smoke (maybe someone already let it out?), no smells, no standby light. Power button didn't seem to do anything.
I decided to look inside, and it was only then that I realised someone had already opened it, because most of the screws on the back cover were missing. An ominous sign perhaps?
Anyway, I took the back cover off and had a look inside (see pic of my ACTUAL board attached). It didn't really look like anything had been touched internally. So either:
- They knew what they were doing so their work is undetectable, and they couldn't fix it :-\ or;
- They replaced entire parts, eg: a whole board, or;
- When the TV broke, the previous owner took a look to see if they could fix it, decided it looked too scary/difficult/expensive to attempt a fix and left it for dead.
I'm hoping it's the latter!
The board details are:
- Model: YP47LPBL
- Part number: EAY60803401
- Revision: 1.12
- Date: 2010/07/05 (Looks like QA was done on 2010/08/28)
You can see some super high detail pics of an earlier revision of this board than mine here, which seems to have an extra relay/contactor thingy compared to my board, but otherwise seems very similar:
Pic 1: https://assets.shopjimmy.com/media/catalog/product/cache/1/image/f8f28dc4a19c21aeb55ee5774693fb5d/s/h/shopjimmy-eay60803401-top_1.jpg (https://assets.shopjimmy.com/media/catalog/product/cache/1/image/f8f28dc4a19c21aeb55ee5774693fb5d/s/h/shopjimmy-eay60803401-top_1.jpg)
Pic 2: https://assets.shopjimmy.com/media/catalog/product/cache/1/image/f8f28dc4a19c21aeb55ee5774693fb5d/s/h/shopjimmy-eay60803401-bottom_1.jpg (https://assets.shopjimmy.com/media/catalog/product/cache/1/image/f8f28dc4a19c21aeb55ee5774693fb5d/s/h/shopjimmy-eay60803401-bottom_1.jpg)
I spent A LOT of time trying to find a schematic for this board but to no avail. Only a crappy tech manual which only seems to have schematics for everything BUT the power supply.
I watched many hours of YouTube vids and found that it's best to first disconnect the power supply board from the main board and test what the low voltage output is when powered up. So I disconnected all cables from the power supply board except the power cord. I've got a fairly basic UNI-T multimeter, but the LV output was pulsing up to around 3.5v. So that made me assume the power supply board was at fault.
Once the incoming AC goes through the first 3x diodes (D100, D101 & D102 - follow the circuit in the 'Pic 2' linked above) the voltage after the last diode is pulsing from 0v up to about 180v at a consistent 1-2 second interval. Not sure if that interval has much to do with the refresh rate of my meter. I also measured a voltage drop across the last diode of around 100v (from memory, haven't tested this for a while). All of the diodes test ok when tested in diode mode.
Tests I've done so far:
- There are no shorts to ground on the LV outputs
- Seems like good continuity between all GND pins and GND on the LV outputs
- Seems to be no shorts to GND on the incoming AC, or on the incoming DC after the above-mentioned diodes.
- Have checked/tested for other shorts around the place and no signs so far.
I've tested most (if not all) of the large capacitors for shorts to ground or being short circuited and they all seem ok. No visible problems with any of the caps! (Not that that means anything).
- I don't have an ESR meter so haven't checked caps in that way yet.
- I've tested all of the large diodes on diode mode and all seem to test ok, although it's been too long since learning electronics at school to know how the zener diodes should test. They all just seemed to work/test like the other diodes.
- I've tested all of the large resistors with 4/5 band ratings. All seem to test pretty much perfect. There were a couple of resistors that I couldn't for the life of me work out the colour coding/rating of (used Digikey's guide). Where these were in pairs/threes, I compared their resistance with other identical resistors and they all seemed to be very closely matched in resistance, nothing out-of-range that would suggest an issue.
Things I've changed so far:
- I thought one of the optocouplers (IC 503) under the centre transformer tested funny/different to the one next to it (IC502), so I ordered a replacement and swapped it. It made no difference, so I swapped the other one too, and still no difference.
- I thought one of the MOSFETs tested weird (in circuit), so replaced both of those (Q102 & Q103). That was fiddly! Even added new thermal paste. But it made no difference.
- I read an EEVBlog post about another SMPS that was pulsing, and someone said that it was probably an issue with the 'snubber' cap/resistor/diode. The resistor (R501) tests almost perfect for it's rating, and the diode (D501) tests fine on diode mode, so I replaced the capacitor (C502), but again, it made no difference.
- I decided to replace any other caps related to the voltage converter chip, and replaced:
- The 5x little 18uf electrolytic caps at C506 - C510
- The 471k ceramic cap at CY 105, and then cos I had lots left, the other 6x of these at CY 101, 102, 103 & 104 near the incoming AC, CY 106 near the pair of transformers, and CY 107 in the corner by itself.
Still no difference!
The only other caps I haven't replaced that, as I learn more I probably should have done as one of the first things :palm:, are the pair of electrolytic caps by the diodes I mentioned at the start and next to the snubber, which are caps C501 & C511. I have replacements for these arriving next week.
The 4x light blue metalised polypropylene caps near the incoming AC all look perfectly fine, and no shorts etc, but again I haven't tested them with an ESR meter. Could consider changing these out for a few $$$.
I was also wondering what the chances were that the voltage converter IC needed replacing? It's an ST Viper27h (see attached pic). Full details here: https://docs.rs-online.com/0422/0900766b813e568b.pdf (https://docs.rs-online.com/0422/0900766b813e568b.pdf) .
That's my next guess if the pair of caps I've ordered doesn't fix it. BUT, that's why I'm posting here, because my guesses so far have sucked haha! Grateful for any suggestions, and feel free to ask any further questions etc.
Thanks for reading my novel... :)
-
Only a crappy tech manual which only seems to have schematics for everything BUT the power supply.
That's not uncommon at all. LG did not design the power supply, it was manufactured by a third party. A lot of companies do this.
- I thought one of the optocouplers (IC 503) under the centre transformer tested funny/different to the one next to it (IC502), so I ordered a replacement and swapped it. It made no difference, so I swapped the other one too, and still no difference.
Quite rare for optocouplers to go bad, and when they do, one of two things will happen: very little to no output from the power supply, or far too much output due to open feedback loop and everything on the secondary side releases the magic smoke.
The 471k ceramic cap at CY 105, and then cos I had lots left, the other 6x of these at CY 101, 102, 103 & 104 near the incoming AC, CY 106 near the pair of transformers, and CY 107 in the corner by itself.
All of those blue capacitors with the "CY" designation actually are not even needed for the power supply to operate. They are strictly there to reduce electromagnetic (radio) interference by providing a path for high frequencies back from the secondary to the primary side. Also note that because these capacitors bridge the low voltage and high voltage sides (which is conveniently indicated by a dashed line on the PCB), they are safety critical, and must only be replaced with the special (often blue) safety-rated class Y capacitors. These capacitors are special in that they are designed so they will not fail short, which would potentially create an electric shock risk if that were to happen.
The 4x light blue metalised polypropylene caps near the incoming AC all look perfectly fine, and no shorts etc, but again I haven't tested them with an ESR meter. Could consider changing these out for a few $$$.
These caps also do not affect the operation of the power supply. They are there to filter out noise (EMI) from the power supply from being conducted out through the AC line cord.
Have you at all measured the 3.5V with the power supply connected to the rest of the TV? Reason why I'm asking, is that some power supplies require a minimum load in order to operate properly, which could be the reason it's pulsing.
-
Maybe this will help... I've taken your picture and identified the various sections of the power supply and what they do.
Ignore every component for now that is not in the standby power supply section, until you've confirmed the standby supply is operating correctly. Again, it may need some load on it to work correctly, what does the 3.5V output measure when connected to the TV main board?
-
Was browsing the forums again and something in the title of your thread caught my eye... you say both the primary and secondary are pulsing?
Unless it's your multimeter's auto-ranging playing tricks on you, that would be a dead giveaway as to what the problem might be: a high resistance path somewhere between the AC mains input and the capacitors C501/C511, so the capacitors charge up to 180V, the power supply starts up but as it draws current and there is too much resistance, the primary capacitors quickly discharge and it shuts down, then they charge back up, and the cycle repeats.
A bad solder joint, a failure of NTC thermistor TH102, or even a blown fuse that is still just slightly conductive, could be possible causes. Heck, for that matter, even a faulty power cord could theoretically cause such a problem.
What I would do is measure the resistance between the live pin of the mains AC input connector and the first rectifier diode, and also the resistance from the AC neutral input to the negative side of C501/C511. The resistance for both paths should be very low, less than one ohm for the neutral path, and around 10 ohms for the live path due to the NTC thermistor. However, for the standby power supply to exhibit a charge/discharge cycling behavior, there would have to be a resistance of well over 1kohm, so it should be fairly obvious, don't worry too much about the exact resistance readings as they are approximate.
-
Hi TheMG
Thanks so much for the detailed responses and info. Very grateful for you going to so much effort!
All of those blue capacitors with the "CY" designation actually are not even needed for the power supply to operate. They are strictly there to reduce electromagnetic (radio) interference by providing a path for high frequencies back from the secondary to the primary side. Also note that because these capacitors bridge the low voltage and high voltage sides (which is conveniently indicated by a dashed line on the PCB), they are safety critical, and must only be replaced with the special (often blue) safety-rated class Y capacitors. These capacitors are special in that they are designed so they will not fail short, which would potentially create an electric shock risk if that were to happen.
Thanks for mentioning this. I of course want this to be safe if/when I get it working. I was quite careful in selecting replacements although I wasn't aware of this point. These are the ones I replaced them with, which were also 471k X1Y1 ceramics, albeit with a higher voltage rating: https://au.rs-online.com/web/p/ceramic-single-layer-capacitors/7368858/
Have you at all measured the 3.5V with the power supply connected to the rest of the TV? Reason why I'm asking, is that some power supplies require a minimum load in order to operate properly, which could be the reason it's pulsing.
I haven't done this yet actually. After reading your later post/points (see below), I get the impression this would be worth considering after trying/testing/confirming the operation of the power supply, right? I can get this put back together so I can test this and report back if you think it's worth doing before addressing the points below.
Maybe this will help... I've taken your picture and identified the various sections of the power supply and what they do.
Ignore every component for now that is not in the standby power supply section, until you've confirmed the standby supply is operating correctly. Again, it may need some load on it to work correctly, what does the 3.5V output measure when connected to the TV main board?
This pic really helps! Especially with no schematic and my limited knowledge of electronics :)
Was browsing the forums again and something in the title of your thread caught my eye... you say both the primary and secondary are pulsing?
Unless it's your multimeter's auto-ranging playing tricks on you, that would be a dead giveaway as to what the problem might be: a high resistance path somewhere between the AC mains input and the capacitors C501/C511, so the capacitors charge up to 180V, the power supply starts up but as it draws current and there is too much resistance, the primary capacitors quickly discharge and it shuts down, then they charge back up, and the cycle repeats.
A bad solder joint, a failure of NTC thermistor TH102, or even a blown fuse that is still just slightly conductive, could be possible causes. Heck, for that matter, even a faulty power cord could theoretically cause such a problem.
What I would do is measure the resistance between the live pin of the mains AC input connector and the first rectifier diode and also the resistance from the AC neutral input to the negative side of C501/C511. The resistance for both paths should be very low, less than one ohm for the neutral path, and around 10 ohms for the live path due to the NTC thermistor. However, for the standby power supply to exhibit a charge/discharge cycling behavior, there would have to be a resistance of well over 1kohm, so it should be fairly obvious, don't worry too much about the exact resistance readings as they are approximate.
This comes in at:
0 ohms from the AC input connector to before the thermistor TH102, and 14.3 ohms after the thermistor and at the input of the first rectifier diode.
The AC neutral input to the negative side of C501/C511 is about 4.4 M-ohms!
I'm looking at the multiple neutral return path routes coming off the negative terminals of C501/C511...and to say it looks complicated is an understatement haha... :palm:
Also, would that neutral resistance be affected by the power supply being powered on vs off?
I'm open to and grateful for any suggestions on how to start figuring this out!
-
0 ohms from the AC input connector to before the thermistor TH102, and 14.3 ohms after the thermistor and at the input of the first rectifier diode.
No problem there, that is right in the ballpark of what I'd expect.
The AC neutral input to the negative side of C501/C511 is about 4.4 M-ohms!
Now you're on to something (or maybe not, keep on reading...).
Also, would that neutral resistance be affected by the power supply being powered on vs off?
No. The standby power supply is always energized as long as there is AC mains to the TV, it provides power to the "brains" of the TV so you can turn it on using the remote control, so that current path should always be near-zero ohms. It's also standard practice to switch the live not the neutral, or to switch both, but there's nothing like that at all happening in this power supply from what I can see.
I'm looking at the multiple neutral return path routes coming off the negative terminals of C501/C511...and to say it looks complicated is an understatement haha...
I see there is something I missed (based on the shopjimmy picture of the back side of the PCB). It looks like the negative side of those caps is in fact NOT connected directly to neutral, it's actually connected to the negative DC output side of the main bridge rectifier, so that the primary high voltage DC of both the standby supply and the main power supply are commoned together. When the TV is turned on and the PFC boost converter is running, power for the standby supply no longer comes from D100/101/102, but instead comes from D10? (blob of silicone over the last digit) adjacent to D601, feeding the standby supply with 400VDC from the PFC converter.
Extra complication for sure, but since it appears the 3.5V supply is responsible for powering the majority of the logic circuits in the TV even while it is turned on, and thus a higher current demand, the manufacturer wanted to make sure it could meet power factor requirements, thus getting power from PFC while on, but directly from half-wave rectified mains while in standby, for power efficiency reasons (albeit at a much poorer power factor). So this design aspect makes sense.
Anyways, what I'm getting at here, is that the 4.4Mohm reading could just be a diode in bridge rectifier in reverse-bias, and not actually an open circuit (a red herring if you will).
I don't know about you, but I'm a visual person, so... I've attached another picture outlining the neutral current flow path, hopefully it helps. Check the path from neutral input to the bridge rectifier, and also check the diode in the bridge rectifier for proper forward/reverse bias reading.
-
Hi TheMG
Thanks again for the follow up and going to the effort with the pics.
I see there is something I missed (based on the shopjimmy picture of the back side of the PCB). It looks like the negative side of those caps is in fact NOT connected directly to neutral, it's actually connected to the negative DC output side of the main bridge rectifier, so that the primary high voltage DC of both the standby supply and the main power supply are commoned together. When the TV is turned on and the PFC boost converter is running, power for the standby supply no longer comes from D100/101/102, but instead comes from D10? (blob of silicone over the last digit) adjacent to D601, feeding the standby supply with 400VDC from the PFC converter.
It's D103 that has the blob of silicone over it (not that it really matters as we have no schematic to refer to). Yes, that was confusing to me how both power outputs seemed to converge. It makes sense now you spelled it out.
Anyways, what I'm getting at here, is that the 4.4Mohm reading could just be a diode in bridge rectifier in reverse-bias, and not actually an open circuit (a red herring if you will).
I don't know about you, but I'm a visual person, so... I've attached another picture outlining the neutral current flow path, hopefully it helps. Check the path from neutral input to the bridge rectifier, and also check the diode in the bridge rectifier for proper forward/reverse bias reading.
This pic, as simple as it is, was actually super helpful haha! I wasn't sure how the bridge rectifier was internally 'wired'. I coulda looked it up, but until now hadn't thought it could be significant.
So I did some tests based on your info:
- From neutral terminal to the bridge rectifier EL115 (through all the chokes) is 0 ohms.
- From negative terminal of bridge rectifier BD101 to capacitors C501 and C511 negative pins is 0 ohms.
- Diode in bridge rectifier measures OL in reverse bias (according to your diagram), and .456v in forward bias.
- Other diode in bridge rectifier measures OL with the positive probe of the meter on the positive terminal and the negative on EL116, and it measures .464v with the meter probes switched over.
- From negative pins of capacitors C501 and C511 to neutral terminal is 5 M-ohms when the meter probes' polarity is correct, and OL if probes are reversed.
Unless I've missed something, this all seems to be testing ok so far and nothing is obviously wrong. Open to any other suggestions :)
-
Hmmm... maybe it's not actually pulsing and it's just your multimeter acting funky? If it is auto-ranging can you try taking the voltage measurements while manually selecting the appropriate voltage range? I've seen some inexpensive multimeters with auto-ranging do some funny things like this.
Other possibility is the TV's power cord or power connector is bad. Check the 120V AC on the two AC input pins of the bridge rectifier if you haven't already. Should be a solid 120V AC no pulsing.
-
Hmmm... maybe it's not actually pulsing and it's just your multimeter acting funky? If it is auto-ranging can you try taking the voltage measurements while manually selecting the appropriate voltage range? I've seen some inexpensive multimeters with auto-ranging do some funny things like this.
This could be the case. My meter is purely auto-ranging and only has modes for AC/DC voltage, and no ranges. Even cheaper meters than this have ranges.
Other possibility is the TV's power cord or power connector is bad. Check the 120V AC on the two AC input pins of the bridge rectifier if you haven't already. Should be a solid 120V AC no pulsing.
We have around 240V AC mains voltage here. I tested between all of the pins on the rectifier and struggled to get more than 2v between any of them.
Between the rectifier pins and GND there was barely any voltage registering, but between the rectifier pins and the L terminal of the connector on the board side was a solid 250v with no pulsing.
The standby power supply area of the board still has around 180V pulsing on AC voltage mode.
The DC output connector is still pulsing from zero to around 3.9v DC.
It's also making a faint 'tick' which sounds like the tick of a clock at about the same frequency as the voltage pulse spikes which seems to be coming from the area where the incoming AC EMI filtering stuff is. Could be one of the chokes/inductors. Really hard to identify exactly where it's coming from. Nothing seems to vibrate if you rest a screwdriver on it or anything.
My replacement caps for C501 and C511 arrived. Is it worth chucking those in and seeing if it makes any difference? They directly/indirectly connect to the areas of the board that supply the voltage controller IC.
Thanks for the ongoing suggestions. Open to any other ideas from TheMG or anyone else :)
-
Solid 250VAC from L to N on SK101?
-
I've got:
- Solid 245VAC from L to N (pads either side of SK101)
- Solid 245VAC from L to SK101
- Solid 0.06V from N to SK101
:-//
-
Are you getting constant DC voltage across C501 / C511? Should be around 330Vdc.
-
Hi Sean
Ok, I think you've cracked the case of the 'pulsing primary AC volts' (we'll get to the ongoing pulsing secondary DC issue later), and this is SUPER embarrassing for me! But I'm gonna be honest... :palm:
Context: At polytech, the tutor said to always first test circuits with your meter set to the highest AC voltage mode to avoid cooking your meter...and then to switch to other modes or DC when you could be sure it was safe to do so...
So the tests I'd done up till now around the standby power supply had been on AC voltage mode, because when I first tested, I got a fairly high and pulsing voltage which I couldn't be sure was being properly rectified to DC.
But after your message, I was like "what the heck", the man asked for DC so I'll test DC. And whadya know!?
I got CLEAN, smooth 340v DC across the terminals of both C501 and C511! Let the insults fly! :-[
I gotta apologise to TheMG for sending them on a wild (Canadian) goose chase. So sorry! :-\ I've updated the title of the thread to reflect what's actually happening.
So after some tests with DCV! I get:
- 340VDC coming out of the first lot of diodes D100 - D102
- 340VDC between GND (SK101) and the primary windings (EL165) of the central transformer
- 340VDC between EL165 and EL166 on the central transformer
- On the secondary of the same central transformer I get 0VAC and 0VDC between EL167 & EL168, and between any of the secondary connections actually.
- BUT I still get the DC output connector pulsing between around 2.9VDC- 3.5VDC at around the same regular frequency of 1-2 seconds as the initial pulsing and as the 'tick' coming from the AC input filtering area.
So now we seem to be getting somewhere, please feel free to make any new suggestions of what/where to test next :)
-
Context: At polytech, the tutor said to always first test circuits with your meter set to the highest AC voltage mode to avoid cooking your meter...and then to switch to other modes or DC when you could be sure it was safe to do so...
That advice is pretty much irrelevant with modern digital multimeters, it was mostly applicable to analog meters.
And yes, I have seen less expensive digital meters have the auto-ranging go berserk and show an unstable reading in AC mode when connected to a large DC voltage (this is actually a common thing to do when you want to measure for ripple on a DC supply voltage). One of the many differences between a cheapo meter and a good meter (never had this problem with the Fluke 289 I use at work).
The good news is, the standby power supply is getting primary input voltage just fine. The bad news: we still don't know if the problem is in the primary or secondary side.
One possibility would be a short or overload on the output causing the power supply to go into "hiccup" mode, but since it's briefly coming up to the full 3.5V as opposed to some lower value, that's probably not the case. Anything getting warm on the secondary side if you leave it powered on for a few minutes?
More likely scenario, is that there is a problem in the primary side circuit that is responsible for providing power to the SMPS controller (IC501: VIPER27), or the chip itself.
The way it works in normal operation is as such:
High voltage present at pins 7 and 8 (DRAIN), the chip allows a small amount of current to flow from this pin through to pin 2 (VDD) which charges up C506. Once the voltage on C506 reaches operating threshold (nominally 14V), the chip begins operation.
After the chip has started operating and is switching the transformer, an auxiliary winding (bottom two pins on primary side of transformer) begins outputting voltage. This is rectified by D502 and charges up C507 via R506. Q503, R505, and zener diode ZD502 form a voltage regulator limiting the voltage being fed to keep C506 charged, this is where the chip gets its operating power VDD during normal operation, after the initial startup.
Aside for the three components Q504, R505, and ZD502 forming an external voltage regulator, this is pretty typical of how a SMPS normally works, and you can also see this in the typical application diagram in the VIPER27 datasheet.
If for whatever reason the charge in C506 is not being maintained from the auxiliary winding, it will quickly discharge until the chip shuts down again, and then start recharging. This cycle will repeat itself indefinitely, leading to a pulsing output as the SMPS starts and stops repeatedly.
Several things could cause this:
-open auxiliary winding, or bad solder connection to it
-D502 open (probably unlikely)
-R506 open
-voltage regulator comprising of Q503, R505, ZD502 not working
I would measure the following next:
-DC voltage on C506: steady/pulsing? what is its maximum reading?
-DC voltage on C507: steady/pulsing? max reading?
-
Hi TheMG
Thought I woulda lost you after the goose chase haha. Really grateful for your patience and ongoing and detailed assistance ^-^
That advice is pretty much irrelevant with modern digital multimeters, it was mostly applicable to analog meters. And yes, I have seen less expensive digital meters have the auto-ranging go berserk and show an unstable reading in AC mode when connected to a large DC voltage (this is actually a common thing to do when you want to measure for ripple on a DC supply voltage)
This is pretty interesting. Wish I'd known! And yes, our polytech tutors got sick of us blowing up their analogue AVO meters :palm:
One possibility would be a short or overload on the output causing the power supply to go into "hiccup" mode, but since it's briefly coming up to the full 3.5V as opposed to some lower value, that's probably not the case. Anything getting warm on the secondary side if you leave it powered on for a few minutes?
It does seem like a sort of 'hiccup mode' in terms of the VIPER chip starting up and turning off like you say, or some similar/related problem. I loved the detailed description of how this circuitry works on this particular board, too. It was great being able to follow it all through.
I did previously have a feel around the board for hot spots but couldn't find anything warm/hot, and nothing looks visually like it's overheated (not that this means much). I don't have an IR camera. I've seen Louis Rossmann splash isopropyl alcohol on boards to see which area is warm by observing where the alcohol evaporates from the fastest. But electronics stores have been sold out of alcohol here for a while as all alcohol was being used to make 'home-brew' hand sanitiser and as a disinfectant ::) I could try again to see if they have it back in stock yet to test for hotspots/shorts.
I also tested the 3.5v output connector for shorts to GND. All of the voltage/signal terminals are at minimum in the K-ohms to the GND terminals, and the GND terminals are 0 ohms to the main GND terminal. So doesn't seem to be an obvious short to GND at least.
Several things could cause this:
-open auxiliary winding, or bad solder connection to it
If I'm right that you're referring to EL166 and its neighbouring transformer connection, this had a 1.2 ohm resistance when I tested it. I re-soldered it but it didn't change. It usually takes about 5 seconds for the meter to gradually climb to this reading for some reason.
-D502 open (probably unlikely)
This measures 0.424v on diode mode in forward and OL in reverse
-R506 open
This measures 21.9 ohms
voltage regulator comprising of Q503, R505, ZD502 not working
Is ZD502 is meant to be installed in reverse? It currently measures OL from positive to negative, and .646v from negative to positive. This is basing the orientation on the fact that it's connected on one side to the negative pin of C507, and this connection has a 0 ohms reading.
R505 measures 1 k-ohm.
Not 100% sure how to test Q503 or if it can be done reliably in circuit.
I would measure the following next:
-DC voltage on C506: steady/pulsing? what is its maximum reading?
Seems to be flickering between 5VDC-13VDC
-DC voltage on C507: steady/pulsing? max reading?
Seems to be almost identical to C506 above and is flickering between 5VDC-13VDC
Both of these caps (C506 & C507) are brand new too.
I'm starting to get excited about this. Seems like we're getting close! Lemme know what to check next :)
-
Is ZD502 is meant to be installed in reverse? It currently measures OL from positive to negative, and .646v from negative to positive. This is basing the orientation on the fact that it's connected on one side to the negative pin of C507, and this connection has a 0 ohms reading.
Yeah it would be a zener diode, so it's normal for it to be reversed biased, it and R505 form a shunt regulator. I'm not 100% sure what Q503 does, maybe over current protection? Someone on here will know.
R505 measures 1 k-ohm.
Probably right? Does it have the value printed on top?
Seems to be flickering between 5VDC-13VDC
Good work, this is the problem, now you just have to find what is causing it. ;)
You've already ruled out some of the possible causes that TheMG mentioned.
Maybe the next step is to have a *really* good look at the soldering, particularly the pins of the transformer and any large through hole components around it.
Could also be the Viper chip is damaged and it's drawing too much current when it turns on, pulling its supply rail too low and making it turn off.
You could try measuring resistance between the VDD pin and Drain and also VDD to GND. You could also try to measure the diode between the GND and Drain pin. But the only way to be sure is to replace it, as it might only break in operation.
-
Yeah it would be a zener diode, so it's normal for it to be reversed biased, it and R505 form a shunt regulator. I'm not 100% sure what Q503 does, maybe over current protection? Someone on here will know.
Actually the resistor, zener diode, and transistor all work together to form a very basic voltage regulator. Zener diode is from base of transistor to ground, resistor goes from collector (input) to base, and the emitter to the load/output. It's an emitter-follower configuration.
The good news is, since such a regulator circuit doesn't allow for reverse flow of current, this means the auxiliary winding, D502, and R506 are doing their thing, otherwise you wouldn't have any voltage on C507, only on C506 getting charged through IC501.
This is getting into the realm of things that are difficult to troubleshoot with just a DMM. An oscilloscope with a high voltage isolated probe (or an isolation transformer), would be able to tell a lot more about what's actually going on.
But it is starting to look like IC501 might be at fault. It's under two dollars on Digikey, might be worth a shot.
-
Hi Sean and TheMG, thanks so much for the messages and suggestions! :)
For R505, and mainly for the transistor, it was super hard to see the markings. But luckily I have a half-decent camera so managed to zoom right in and get a close up photo for a better look (see attached pic).
R505 is marked '102' which seems to mean it should be a 1 k-ohm resistor, so that looks good.
As for the transistor, there are only two characters visible on it ("PZ"?) and I'm guessing there should be a number as well? Even with the original full-resolution image, I can't see any other figure. Is there any way to be able to work out what this should be based on the surrounding components and test results? :-//
Maybe the next step is to have a *really* good look at the soldering, particularly the pins of the transformer and any large through hole components around it.
Did this. Re-soldered the all of the transformer joints, and also re-soldered the large through-hole stuff nearby, including the pins of the Viper chip. I just powered up the board again to test it and the soldering hasn't made any difference. The DC output is still "hiccup-ing" between 2.5VDC and 3.5VDC.
You could try measuring resistance between the VDD pin and Drain and also VDD to GND. You could also try to measure the diode between the GND and Drain pin. But the only way to be sure is to replace it, as it might only break in operation.
- VDD to either drain pin = OL
- VDD to GND pin = OL
- VDD to negative of C507 = 17 M-ohms
- GND pin of Viper to GND of board = OL
- GND pin of Viper to negative of C507 = 0 ohms
Please let me know if any of the above readings seem out of range.
But it is starting to look like IC501 might be at fault. It's under two dollars on Digikey, might be worth a shot.
Happy to replace the Viper chip if that's the next logical step. Thanks so much for the help. I'll be sure to update when the chip comes in and is installed :D
-
Ok, so my Viper chips arrived today (had to order a minimum of 10 from local supplier, which was still cheaper that postage to Australia for a single chip from Digikey)...
I soldered a new Viper chip onto the board, anddddd... *drumroll* ...it's still doing the same thing! :palm: This has got me stumped!
Something about me which is a blessing and a curse is I'm pretty persistent. I still don't want to give up on working this problem.
But I'm already beyond my current electronics knowledge. All I'd really be able to do now is to start doing more 'shotgun' approach (changing everything until it starts working).
I'd be really grateful for any more ideas/suggestions of what to look for/change/do next to keep troubleshooting :)
-
How about checking capacitors C204, C205, C506 to C510?
-
Hi shakalnokturn
Thanks for offering some suggestions! :D
For capacitors C506 t0 C510, I have actually already replaced all of these with brand new capacitors of the same value and unfortunately it didn't help.
But I haven't yet tested/replaced capacitors C204 and C205. According to one of TheMG's first diagrams, these are also part of the 3.5V power supply, so your suggestion to investigate them makes a lot of sense. I will check these out next and report back. Much appreciated :)
-
Sorry about C506 to C510... The topic was already quite long , I did fly over some of it.
-
Hey team. I found something of interest when testing shakalnokturn's suggestion.
Both caps C204 and C205 actually seemed ok from some basic multimeter tests (I don't have an ESR tester). But both had NO shorts to ground, zero ohms continuity to GND on their negative terminals, and almost identical resistance internally. If one was standing out as having a noticeably different reading from the other, I'd probably be more concerned about them, but they seemed ok. I also don't fancy waiting another week for them to arrive, as I can't get them locally. Especially if they don't make any difference haha!
But I decided to test the other components around the VIPER chip. I tested the small SMD caps, but they seemed ok based on same tests as the caps above.
Then I looked at the SMD resistors coming off the VIPER chip. The first one I tested was R502, marked 474 (470k-ohms, see pic). It turns out it's reading just 10.2ohms! If I understand what you guys have been saying previously, could this cause the capacitors that help the VIPER chip to start-up to drain before the chip has completed its startup? Then cause the startup loop 'hiccup'? :-//
I want to replace this resistor before anything else as it's WAY out of spec. But I have three choices of 470k SMD resistors available locally and I have no idea how to calculate/choose which one to use, OR if any of them are going to be any good and if I need something else.
The 3x options I have locally are as follows, they all have this same package, but have different wattage's (option listed below): https://www.altronics.com.au/p/r1268-470k-.125w-805-metal-film-smd-resistor-pk-10/ (https://www.altronics.com.au/p/r1268-470k-.125w-805-metal-film-smd-resistor-pk-10/) (picture on listing is indicative so markings on actual item may be different)
Should I choose:
0.125W
0.25W
1W
Or something else? :-\
If it helps, you can get the full VIPER chip specs and typical circuit etc here: https://docs.rs-online.com/0422/0900766b813e568b.pdf (https://docs.rs-online.com/0422/0900766b813e568b.pdf)
Grateful for any suggestions :)
-
Sorry, I could not follow all of it properly as I am pretty ill right now. But I did see something like this once before. The pulsing power supply is caused by it trying to close down as soon as it starts up. Interesting that it seems to be once the logic gets going properly. What I came across was that it was getting a signal to turn itself off from the remote control board. Error on the board, leaky IR diode and don't forget potentially a sticky "Power" button on the remote, or the front of the TV.
I hope I have been able to help in some way, after all the work on it you have done, you deserve to get this one going!
-
99.9% of the time, resistors fail high resistance or open-cicrcuit.
The chances of a 470k resistor going to down to 10 ohms is pretty much non existent. If you are measuring it while it's in the circuit, it's probably something else in the circuit that you're actually measuring not just the resistor itself.
-
Appreciate the suggestions Fred Basset & TheMG
Fred: From what you said, I've hopefully eliminated all/most of those issues as I've been working/testing/powering up this power supply board outside of the TV and it's not connected to the main board controller which connects to the buttons, IR receiver etc. I also don't have a remote for it :'( Let me know if I misunderstood you. Those are still useful tips for future troubleshooting and I wouldn't have considered them had you not mentioned them, so thanks. By the way, feel better soon!
TheMG: I'd heard the same in that resistors usually fail in the manner you said, 99.9% of the time. But when a resistor you're working on fails in a different way, that 0.01% becomes 100% to you, right? :-DD
10ohms is much closer to open circuit than it is to 470k-ohms. Yes, it was in circuit when I tested it, and meters/electronics can do some weird things and give strange readings at the best of times. I guess I'll see if I can ping it out of circuit and test it again. For the ease of replacing that resistor, it's surely worth a try, right? :D
-
Ok, so I took resistor R502 out of circuit and tested it. What a tiny fiddly little thing!
So test results were:
- Still around 10 ohms between the pads that the resistor connects to on the board AFTER the resistor was removed!
- The resistor itself measured .468M-ohms...so basically 470K-ohms as it's supposed to.
Sooo...TheMG was right! And I was silly for thinking I knew better :-[
I found another SMD resistor next to the capacitors C204 & 205 that measured OL in-circuit. So after my above schooling in electronics, I took that out of circuit and tested that too, and it was bang on 47ohms as it should be. This is teaching me a lot about testing these circuits.
UPDATE: Weirdly, with the second SMD resistor mentioned above (R202), after I soldered it back to the board, it now measures 47 ohms in circuit!? I tested it a couple of times before I took it out of circuit too because I wanted to be really sure that it was OL before removing it. So not sure what that's about. This made me think that perhaps it had a crappy solder joint before I touched it. So I powered up the board to test it, but it still has the issue with the secondary output 'hiccuping' from around 2.5VDC to 3.5VDC.
Is there a more scientific approach I can take to figuring out this fault besides just replacing stuff?
Should I still order replacements for C204 & C205? Any other suggestions on what to try next?
I really want to solve this puzzle! Will be so satisfying after all of this nonsense! :)
-
Ahh...Thanks for the kind wishes, I appreciate them. Just sorry I was not able to help. I once spent quarter of an hour trying to diagnose an intermittent fault on a TV only to find out it was a rubber membrane button had got stuck under the keyboard fascia.
When the IR receiver starts going out of tolerance, you can have all kinds of fun like the TV turns off whenever someone turns the room light on, or the color goes out of the picture at the same time every day (it is reacting to the sun coming through the windows).
Good luck, I hope you can nail it! The only way to scientifically nail it is usually with a full service manual and sometimes a lot of thinking. Do you have an oscilloscope and/or a microscope? Even a magnifying glass would help. Look for bad joints (may be partially hidden) and any possible discontinuities in the tracks. Unfortunately that could be in any power rail.
Oscilloscope is handy for checking you have the correct waveforms around the VIPER chip and correct voltage levels (no noise or oscillations). Without these, you might be best keep replacing the most likely parts. If the part is fine, then keep it in your stock for your next repair job and consider it an investment! That is what I tell myself.
-
might be a good time to look at the psu outputs instead. The first thing to do with a pulsing power supply where it is hard to take readings is to disconnect an output like the backlights to see if it settles down and stabilizes. A faulty board on the output will load down the psu and stop it functioning correctly and often cause pulsing issues. The no stand by light is odd tho and may be a secondary issue or no issue at all depending on a number of other factors. Disconnect the backlights up the top of your board and see what happens.
-
might be a good time to look at the psu outputs instead. The first thing to do with a pulsing power supply where it is hard to take readings is to disconnect an output like the backlights to see if it settles down and stabilizes. A faulty board on the output will load down the psu and stop it functioning correctly and often cause pulsing issues. The no stand by light is odd tho and may be a secondary issue or no issue at all depending on a number of other factors. Disconnect the backlights up the top of your board and see what happens.
Hi 'login'. I don't blame you for skimming/not reading the full thread as it's getting long, or for assuming from some of the pics that these tests are being done with the PSU board still connected to the main board and backlights etc. But from the beginning of this thread the PSU board has been disconnected from everything (except the power cord when powered up, obviously), and I'm still getting the pulsing on the 3.5VDC output of the PSU board. Appreciate the suggestions all the same. :)
The only way to scientifically nail it is usually with a full service manual and sometimes a lot of thinking. Do you have an oscilloscope and/or a microscope? Even a magnifying glass would help. Look for bad joints (may be partially hidden) and any possible discontinuities in the tracks. Unfortunately that could be in any power rail.
Hi Fred. Unfortunately:
- I can't find a manual for this board. I've searched high and low. I can only find a manual for the main control board.
- No oscilloscope :( If I was a regular hobbyist I could justify it. I'm also about to relocate soon so I'm trying to acquire as little stuff as possible :-\
I think I'll just have to keep manually tracing things with the meter. Is it worth re-flowing virtually every single solder joint? :-//
-
oh ok my bad. you are right I did kinda skim the thread and didn't pick up that piece of info. It wouldn't hurt to resolder the entire board or at least around anything that generates heat as you can not always see a dry joint and it may look ok but isn't. Good luck
-
That's a good tip, 'login'. Thanks. Will get into it over the weekend. :)
-
I had a suspicion you did not have a scope and could not get the manual. Just try to think of this as a good learning experience.
Another old trick is a hairdryer to heat the board until the fault devolps, then a hot soldring iron to narrow it down to component level. In this case we have the opposite problem of it not working. The usual suggestion is freezer spray then to see if you can get it to behave any better. This idea being that the localised contraction will hopefully make any dry joints or cracks in the board close enough to give you some idea of where the problem is.
It is mind-numbing, but when all else has failed me, I use a meter on continuity to check all the tracks from pad to pad and track to track to see if the readings make any sense?
-
I found another SMD resistor next to the capacitors C204 & 205 that measured OL in-circuit. So after my above schooling in electronics, I took that out of circuit and tested that too, and it was bang on 47ohms as it should be. This is teaching me a lot about testing these circuits.
It might have been that the solder was oxidized or had dried flux etc, sometimes the probe tips don't cut through it making it measure OL. Or maybe it was a solder issue, hard to know.
BTW do you have all the outputs disconnected? That would rule out an issue with the backlight etc.
-
It is mind-numbing, but when all else has failed me, I use a meter on continuity to check all the tracks from pad to pad and track to track to see if the readings make any sense?
I did a bunch of this along with re-soldering some things in the last couple of days. My mind is getting seriously numb! :palm:
It might have been that the solder was oxidized or had dried flux etc, sometimes the probe tips don't cut through it making it measure OL. Or maybe it was a solder issue, hard to know.
BTW do you have all the outputs disconnected? That would rule out an issue with the backlight etc.
Hi Sean. I hear your point about the probe tips not cutting through the old solder. Even when testing some points in the last day or two, I need to really dig the probes in sometimes, or test multiple areas of some pads to double-check a reading.
I've mentioned previously that I have all of the outputs disconnected from the board to eliminate any issues from beyond this board.
After re-soldering a bunch of points, the board is still giving the 'hiccup'/pulse on the 3.5VDC output going between 2.5VDC and 3.5VDC.
I get 330VDC going into the drain of the VIPER IC chip (IC501).
The central transformer which supplies the drain of the VIPER IC chip has no voltage between any of the terminals of its secondary (The terminals including and next to EL167 & EL168). It also has no voltage between those secondary terminals and N/GND.
I get the impression that something attached to/downstream of the VIPER IC is broken and is stopping the VIPER from booting up properly. I have pretty much tested all around the VIPER IC.
I just ordered replacements for capacitors C204 & C205 as these are the only electrolytics in this area which I've not yet changed. Visually they look fine and they both have almost identical multi meter resistance and are not shorted, or shorted to GND. But 'shakalnokturn' suggested it earlier so what the heck!
How can I identify replacement parts for the tiny SMD capacitors/diodes/transistors next to the VIPER IC? (See pictured). Some have markings, but the SMD capacitors don't (and I don't have a schematic diagram which really sucks!). Even the components which are marked (like the resistors), how can you tell which current/voltage ratings to use?
Please feel free to make suggestions, or ask me to test things and report back with results etc. Thanks! :D
-
Are you still using the replacement VIPER IC, or did you put the original one back?
Hiccup mode is not the only way of shutting this thing down though, it is pretty good at closing itself down to protect things.
so I am now thinking what if something was triggering the control pin? Like high voltage tracking across the board. This might not show with a test meter and the voltage off. With the power supplied, the pin would go too high and shut down before it could be measured with a multimeter.
It also has brownout protection, so anything dragging it down would trigger it.
I am rapidly running out of ideas, anyone else can think of anything?
My personal feelings are that the voltages are the problem, but if they are reading ok with a meter, then it is interference on the voltages - something that is not showing up on DC readings.
Full credit to you though for sticking with this, I think a lot might have been tempted to dump it before now as taking up more time than it is worth. Unfortunately now I am curious as well.
-
Hi Fred. I'm still using the brand new VIPER IC. I figured I'd leave the new one in. Because changing it made no difference to what the board was doing, so it likely wasn't the VIPER IC that is causing the issue.
Hiccup mode is not the only way of shutting this thing down though, it is pretty good at closing itself down to protect things.
so I am now thinking what if something was triggering the control pin? Like high voltage tracking across the board. This might not show with a test meter and the voltage off. With the power supplied, the pin would go too high and shut down before it could be measured with a multimeter.
It also has brownout protection, so anything dragging it down would trigger it.
Yes. It seems very much like the VIPER IC is shutting itself down due to some sort of issue up/down stream. I was just looking at the VIPER IC spec PDF (https://docs.rs-online.com/0422/0900766b813e568b.pdf (https://docs.rs-online.com/0422/0900766b813e568b.pdf)). I did some live testing on the VIPER chip the other day. I don't remember exact values, but I'm pretty sure there wasn't more than 8.5VDC measured on any of the control pins. And from memory, this was cycling from around 4VDC or so up to that level. The drain had the full 330VDC ready to go.
Would doing some more tests on the VIPER IC and reporting back the results be helpful? If so, please specify which tests. Thanks :)
-
Would doing some more tests on the VIPER IC and reporting back the results be helpful? If so, please specify which tests. Thanks :)
At this point the best tool to use would be unfortunately the one you don't have: an oscilloscope.
Whatever is happening, is happening too quickly for the multimeter to tell the whole story. For example, is the IC shutting down then starting up because the VDD keeps dropping too low? Or is the IC shutting down for some other reason, and the VDD is dropping because the chip shuts down? Which happens first? With an oscilloscope you could see that, but you'd never be able to tell with a multimeter.
-
I am glad the new VIPER chip is still in. I was thinking along the lines of multiple faults. Say the chip failed, damaged something and so the new chip made no difference, but then you put the old chip back so we could not find a fault around it, because the chip was faulty as well.
Sorry, but the only ideas I have right now can only be sorted with an oscilloscope. I even considered slowly changing waveforms that you might be able to detect on an old analog meter. But this seems to be a quick thing that is messing the chip up. Are you SURE you don't want to buy an oscilloscope? Just think of all the things you could fix in the future with it...
-
At this point the best tool to use would be unfortunately the one you don't have: an oscilloscope.
Sorry, but the only ideas I have right now can only be sorted with an oscilloscope.
Ugh :palm:
Thanks for the help TheMG and Fred. I hear ya! And boy do I wish I had a 'scope to be able test what's going on. I'm just not in a position to get one now as I'm moving soon, and I also don't know how much I'd even use it. I guess I could probably find one nearby that I could borrow.
I can actually get a replacement board pretty cheaply on Ali Express (risky!), but that's nowhere near as satisfying as fixing the one I have. Plus it will likely take weeks to arrive unless I wanna pay a crazy amount for shipping.
I've got those C204 & C205 replacements arriving soon. If they don't work...I dunno...I still don't want to quit. I'm learning so much about electronics working on this. I would be pretty disappointed if I don't get to fix it and need to just throw it away when I move :(
-
Keep us updated on those two electrolytics before you move. You've been through most of the standby PS by now.
You'll be wanting to get the TV sorted before you move, failure to do so is the beginning of the end. In a few years time you'll be dragging heaps of junk "to be repaired when..."
If you're enjoying this repair attempt, struggle through it as far as you can, buy a scope once you've moved you'll realise how much easier it is when you get a glimpse of what's happening.
-
Thanks shakalnokturn
I know you want to know what happens with the C204 & C205 capacitors more than anyone, 'cos if it works, it was your suggestion, so you get the glory :-DD
I'm teasing haha ;) I hope they work! I'll definitely keep you posted. Hopefully it will be good news!
-
I mainly like topics to be updated and when possible with a happy end.
I don't mind a little glory when it doesn't cost me too much ;D
Considering the symptoms and it being the standby PSU (powered all the time for years) the filter capacitors on secondary side and VIPER Vcc primary side filter cap. would definitely be my starting point.
The problem is without an adequate way of testing you can either run in circles or have to swap just in case.
One of those AVR based component testers is a good investment too, although not the best tool around for the job (great Swiss-knife though) it would give you a better idea on capacitors condition than a multimeter.
-
Well my C204 & C205 replacement capacitors arrived today. Soldered them in. Tested, and...
Still doing the 2.5 - 3.5 VDC 'hiccup' on the output connector :palm:
I took some measurements at the VIPER IC chip, although as you guys said, they're form a digital multimeter and probably worthless:
- From GND pin of VIPER IC to the board GND is moving between -150mvDC to -190mvDC
- From VDD pin of VIPER IC to the board GND is moving between 4VDC to 12VDC
- From CONTROL pin of VIPER IC to the board GND is moving between .3VDC to .35VDC
- From FB pin of VIPER IC to the board GND is moving between .3VDC to .5VDC
- From Drain pins of VIPER IC to the board GND is solid at 330VDC
The secondary terminals of the central transformer are all still at 0VDC to GND of board.
Open to any other suggestions... :horse: :)
-
When you test do you test in the TV with everything wired or just PSU on the bench?
-
:-DD :-DD :-DD time to try another forum m8 b4 you end up replacing every part in the power supply. It was never going to be the caps.
-
When you test do you test in the TV with everything wired or just PSU on the bench?
So far have just been testing with PSU on the bench, with everything DIS-connected. The reason I haven't tried connecting it back into the TV is because it's not exhibiting any different behaviour on the bench to warrant connecting it back to the TV for testing. If you believe I ought to try it connected back in the TV, please tell me your reasoning for doing so as well so I can understand the methodology.
:-DD :-DD :-DD time to try another forum m8 b4 you end up replacing every part in the power supply. It was never going to be the caps.
Perhaps. Not sure if you've read the whole thread (don't blame you if you haven't), but out of respect to the other contributors, caps aren't the only thing that's been suggested/attempted. If "it was never going to be the caps", what IS it going to be, 'login'? And what other forum are you suggesting I try?
By the way, a test result I forgot to post yesterday is the C502 capacitor in the 'snubber' circuit has 340VDC between either terminal to the board GND, but between it's terminals, it is alternating between 340VDC and 0VDC every 1-2seconds.
-
I was asking about test conditions because they can affect the outcome.
Testing in TV excessive loading from a fault on mainboard could have put the standby supply in an over-current condition.
OTOH some PS designs need a minimum load to regulate correctly. There is no on-board load on this design, until the mainboard has activated power-on line, after that standby power supply is loaded by mainboard, remote receiver, relays on PSU it must also supply PFC and main PS IC's.
With all that has been checked and replaced I'm surprised that nothing is better. It may be worth testing with a 2W 10-20 ohm resistor on output.
Sorry if this has been asked before, I haven't re-read the full topic: Did you test T501 for continuity on winding linked to D502 and primary ground?
-
Hi shakalnokturn
Thanks for explaining the thinking behind the suggestion about trying the board connected vs disconnected. Someone did previously mention that some boards need a small load to be able to complete their start up. Imagine that this PSU board was fine the whole time, but the main board on the TV was faulty and that was the issue!? :wtf:
I'll see if I have a resistor lying around in the range you suggested which I can use for a test. Do I just bridge one of the 3.5VDC output connector terminals to one of the GND terminals? If I can't find a resistor, I could just try a test with the board in the TV and hope no smoke is released :-DD Or go buy a resistor for a few pennies...
Sorry if this has been asked before, I haven't re-read the full topic: Did you test T501 for continuity on winding linked to D502 and primary ground?
No one has specifically suggested testing the continuity along the T501 winding between primary GND and up to D502, but TheMG did suggest testing the windings and transformer connections, which have all been re-soldered.
I also did actually test the continuity in this part of the board the other day too when tracing the circuits around the VIPER IC501.
The readings are:
- Transformer winding EL166 to primary GND = Reading gradually climbs for about 5 seconds from .1ohms to 1.1ohms
- Through the same winding to the other side of D502, testing in diode mode in forward direction for diode = .449v, and OL in the reverse direction.
Please let me know if these results seem within range to you as well :D
-
I'm out of ideas as well, I'll need to think a bit more.
One thing to be careful of, some TV power supply boards must be connected to the TV chassis, the screws and the body of the TV complete the GND circuit. Powering them up on the bench can destroy a lot of components on the board. But it looks like this board isn't one of them, but something to keep in mind if fixing other TVs. ;)
-
Thanks, Sean. That's a good tip. I'll try hooking it all back together and see if anything happens...or smokes... :D
-
The readings are:
- Transformer winding EL166 to primary GND = Reading gradually climbs for about 5 seconds from .1ohms to 1.1ohms
- Through the same winding to the other side of D502, testing in diode mode in forward direction for diode = .449v, and OL in the reverse direction.
Seems ok to me, although I'm a little unsure of the pinout of the primary side of the transformer.
Did D501, D503, ZD502, ZD501 and D201 (on secondary) test okay too?
The zener diodes (ZD) can't be tested completely with a multimeter, because they should have a reverse breakdown at some voltage. But if they test 0.6V forward and OL reverse they are *probably* ok, it rules out that they are shorted.
I guess it's also possible that the other diodes block in the reverse direction for low voltages (from the multimeter) but might breakdown and stop working at higher voltages when the TV is turned on. I'm not sure how common this failure mode is though? Replacing the diodes would be the easiest way to check.
Thanks, Sean. That's a good tip. I'll try hooking it all back together and see if anything happens...or smokes... :D
Yeah it gets a lot of people because it's weird for circuit boards to self destruct if powered up outside of the equipment. :-BROKE
But with your board it looks like all the screw holes that are joined to circuits are connected together on the board, so it should be fine to power up on the bench (but it might need a load, hard to know).
-
Did D501, D503, ZD502, ZD501 and D201 (on secondary) test okay too?
All tested ok in diode mode. Noted on the zener diodes. BUT, I haven't removed D201 and tested it out of circuit. Thought this was a transistor.
Yeah it gets a lot of people because it's weird for circuit boards to self destruct if powered up outside of the equipment. :-BROKE
But with your board it looks like all the screw holes that are joined to circuits are connected together on the board, so it should be fine to power up on the bench (but it might need a load, hard to know).
Ok...so I connected the board back into the TV. So far I've only connected the main board back and not the LED lights. Now I'm getting a solid 3.5VDC (3.49 to be exact) on the output connector, and I'm getting 3.5VDC over at the main board connector. Sooo...it's looking like the power supply coulda been fine the whole time, but needed a load connected... :-[
Still doesn't seem to be a standby light on. Geez. This is upsetting... :palm:
Now I gotta try troubleshoot the main board... The most common 'advice' I've seen in the past on these main boards is to 'cook' them in the oven :-// Is there another/better way of troubleshooting/fixing these and verifying they are working properly?
-
I decided to connect up the LED light connectors to the main board too in case they are required for the TV to know everything is OK and for it to power on.
The TV is lying face-down so I can work on the boards on the back. As I lifted the TV to check for the standby light with all of the sub-boards connected and with the TV plugged in and powered on, I heard a beep and a click. It made me jump! :-DD
I'd accidentally pressed one of the control buttons (capacitive buttons? They are behind the front panel glass and you just press on the glass). So I then pressed a bunch more of the buttons. All of the buttons which would cause the TV to power on (eg: Power/Menu/Program +/- etc) cause it to beep and the relay RL101 to click. It sounds like the relay clicks on and then almost immediately clicks off again. So perhaps a fault condition of some kind?
There is still no standby light visible.
When I first found the TV and plugged it in, I tried pressing the power buttons and nothing happened. So it seems weird that it's beeping/clicking now. But now I gotta work out:
- Why the relay seems to click on then off right away
- Why there's still no standby light
- Why it won't turn on
Progress! (Sort of...)
-
BUT, I haven't removed D201 and tested it out of circuit. Thought this was a transistor.
I think that should be okay to test in circuit, did you still get 0.6V forward and OL for reverse?
Ok...so I connected the board back into the TV. So far I've only connected the main board back and not the LED lights. Now I'm getting a solid 3.5VDC (3.49 to be exact) on the output connector, and I'm getting 3.5VDC over at the main board connector. Sooo...it's looking like the power supply coulda been fine the whole time, but needed a load connected... :-[
Nice, that is progress. At least you didn't buy a new power supply board. ;)
Which board does the standby light come from, the power supply or the main board?
-
I think that should be okay to test in circuit, did you still get 0.6V forward and OL for reverse?
For D201, please excuse my ignorance as for what the pins are. All I can tell is there are 2x short pins and 1x longer pin :D I've attached a pic showing the test results for it IN-CIRCUIT. The arrows show the direction of the test in diode mode, so the arrow head is the negative lead, and the arrow tail is the positive lead.
Nice, that is progress. At least you didn't buy a new power supply board. ;)
Which board does the standby light come from, the power supply or the main board?
Yes! Very pleased I didn't buy another PSU board. Woulda still been weeks away from arriving too!
The standby light seems to come from the mainboard (see attached pic).
-
For D201, please excuse my ignorance as for what the pins are. All I can tell is there are 2x short pins and 1x longer pin :D I've attached a pic showing the test results for it IN-CIRCUIT. The arrows show the direction of the test in diode mode, so the arrow head is the negative lead, and the arrow tail is the positive lead.
So the two shorter ones are joined together on the PCB. Looking at the pic of the underside of the board it looks like the part number might be SBRF2060CT, have a look at the datasheet for that, I think the diagram in the datasheet will make it clearer.
The 0.146V seems a bit low for the forward voltage, but maybe it's normal, someone on here will know.
I'm also not sure why the reverse voltage is 0.786V. Does it measure OL if you unplug the main board? What is the resistance from the diode cathode (long pin) to GND (nearby screw hole) with and without the main board attached?
Also, thanks for posting the pic of the back of the TV, that makes it clearer. ;)
edit: The diode is probably okay because your getting the 3.5V, but the measurement values don't seem quite right, might be indicating a fault on the main board.
-
Thanks for the assistance, Sean.
So the two shorter ones are joined together on the PCB. Looking at the pic of the underside of the board it looks like the part number might be SBRF2060CT, have a look at the datasheet for that, I think the diagram in the datasheet will make it clearer.
It did, thanks for finding this.
The 0.146V seems a bit low for the forward voltage, but maybe it's normal, someone on here will know.
This remained the same even with the main board disconnected.
I'm also not sure why the reverse voltage is 0.786V. Does it measure OL if you unplug the main board? What is the resistance from the diode cathode (long pin) to GND (nearby screw hole)?
- Yes, the reverse voltage did change to OL once the main board was disconnected.
- The resistance of the cathode to GND is .791K-ohms regardless of whether main board is connected/disconnected
-
By the way, here's repair manual which may provide some clues. There's no schematics for the power supply, but there are some schematics for some of the power and control circuits on the main board.
-
Here's a pic of the main board
-
Here's a pic of the power supply board in case anyone wanted to trace the control circuits for the relay RL101
-
If your 3.5V supply is stable during the relay clicking sequence I'd call it good.
If the clicking is triggered by a key press there's some hope for the mainboard (don't cook it blindly, a good way to kill it without necessary need).
Do you get the slightest blink of backlight while the clicking sequence happens?
Do you catch any voltage on the 20V output during the clicking? (Don't slip while measuring, shorting the 20V and 3.5V supply's could be counter-productive.)
At this point I'd probably lift the Power-On line from one of the connectors to disconnect the mainboard control and tie it to 3.5V on the PSU side and check that the main supply is working as intended.
-
Hi shakalnokturn
Do you get the slightest blink of backlight while the clicking sequence happens?
I didn't notice this, but I didn't proactively look for it either. Will see if it happens next time I test it, and will try and do it in the dark so it will become more visible.
Do you catch any voltage on the 20V output during the clicking? (Don't slip while measuring, shorting the 20V and 3.5V supply's could be counter-productive.)
The relay on-off click is pretty fast. It's like the relay clicks off immediately after it clicks on, kind of like a heartbeat rhythm. I'll have to carefully test this and will report back.
At this point I'd probably lift the Power-On line from one of the connectors to disconnect the mainboard control and tie it to 3.5V on the PSU side and check that the main supply is working as intended.
I'm not 100% sure what you mean by this, shakalnokturn. Are you suggesting I disconnect the mainboard from the PSU, and then put a jumper between the Power-On pin and the 3.5V pin as per the attached pic? :-//
-
Apparently the power supply does like a little load so either power it on the bench with a resistor to load the standby supply and force power on by linking as suggested on your photo or just isolate the "Power-On" from the connector (to not damage mainboard by sending 3.5V when it is holding it low) using a needle to release it and force active as previously.
You should be able to check if you get the 12 and 20V outputs at least. You probably won't get the backlight as that needs controlling separately.
I'd go for the second option as you're actually supplying the mainboard as a load that way, if the 12V and 20V outputs appear low check that you're getting around 400V across the primary capacitors C611-C618.
If that's good check value of dark red capacitor C114.
-
i will sit this one out as I think it is a time waste with a lot of guessing involved and probably looking at the wrong board.
the common faults with this model are backlights and main board. The common problem with the main board is the solder balls under the BGA chip dislodge and need to be reseated. The professional way to repair that is thru a service center with the proper equipment to do the repair properly. Throwing the board in an oven is a crude, primitive way that is a temporary fix at best. It may work it may not but usually is not permanent.
Some forums have a dedicated tv repair section and are easily googled. It's probably not fair on those trying to help you to steer you away from here just yet. If you hit a wall here I'll tell you then.
-
If you are familiar with LCD TV repairs and know of the usual faults on this model maybe you could give some input on how to go about doing a proper diagnosis with minimum tools.
We have no spare PCB's from another TV but we have a DMM to measure voltages, what would your approach be to test the mainboard?
Please contribute even if your method involves tools that are not available I'm curious to learn the way things should be done in the trade.
-
Some forums have a dedicated tv repair section and are easily googled. It's probably not fair on those trying to help you to steer you away from here just yet. If you hit a wall here I'll tell you then.
You mean the Badcaps forum? That's a really good site, OP could start a second thread on there if they want to. ;)
-
The passive aggressiveness here is amusing. We all gonna meet behind the bike sheds after school to 'throw down'? :box:
'login' has some valid points. When I first looked up these TVs (LG47LE7500, and related models of LG LCD LED TVs), there are common problems with the main boards and back lights as 'login' said. Often the BGA chips' solder joints deteriorate, casing intermittent or other strange faults. I've not really seen another one exactly like this, so it threw me.
When initially testing the TV, I saw methodologies on forums and online which suggested disconnecting the power supply board from the main board to be able to properly diagnose if the main board or power supply (PSU) was at fault. Through inexperience and ignorance, I didn't realise that the PSU needed a load connected (the main board) to actually have a stable standby voltage at the output connector (among all the other things I didn't realise and have learned from the generous contributors on this forum).
As I stated from the beginning, I'm new to electronics, and was doing this for the fun/learning. I'm not running a commercial TV repair operation (LUCKILY! I'd be broke otherwise haha! :-DD )
Anyway, now much of the initial nonsense/learning is out of the way, it's starting to look more and more like the classic main board problem these models experience. Yes, the ideal thing to do is to remove the main chip and do a full, proper renovation of the BGA solder joints. I could probably do this well enough after a bunch of attempts. There are plenty of tutorials on YouTube with varying degrees of professionalism, ranging from fully professional with all of the proper equipment, with proper chip heaters and digital temperature profiles etc, all the way to people using heat guns to mimic the temperature profiles to reflow/flow their chips, AND, as I mentioned earlier, even people who literally oven-bake their main boards to get that oven-fresh plastic and chemical smell in their own home. The latter two methods are of course not the right way to do it, and it comes with the 20/20 warranty (20 seconds or 20 steps, whichever comes first).
I paid $0 for this TV. It's a 'dumpster dive special'. As much as I'd like a fully equipped electronics lab, right now, I don't have the time, budget or space for one. I also have a point where this learning experience starts to cost more than I'm prepared to pay. Because if I get this TV working, I have no need of it. I'll probably either give it away or sell it. It's not really worth it to me to spend more than the value of the TV in trying to fix it.
So (to 'login's' disgust), I'll probably next try something akin to a dodgy reflow of the main BGA chip, using none of the correct tools, because I don't have them. BUT! I'll splash out and treat the board to some flux during the process.
If I destroy this board, I'll be disappointed, but I'll be no worse off. And maybe it'll even work! Will update this thread with what happens...
-
the common faults with this model are backlights and main board. The common problem with the main board is the solder balls under the BGA chip dislodge and need to be reseated.
LG models around this era are notorious for BGA joint issues, I've repaired this same model (47LE7500) and a 60PZ550 60" plasma by removing and reballing the BGA. The symptoms for the 47LE7500 were no picture but had sound (had to blindly select the HDMI input). You can add flux and reflow the chip as a test to see if you get any difference in functionality to narrow down the issue.
Edit: Just noticed Fixin Stuff replied as I was submitting this reply. Only note I will add is that if the oven reflow works, it's a temporary fix and will likely fail again so I wouldn't recommend selling it absent warnings that the set will likely die in a few months.
-
Hi 'taligentx'. Well done for fixing one of these before. Yes, the BGA issues all seem to manifest in weird and wonderful (and often unique) ways depending which of the solder balls are the worst. I like your suggestion of trying a reflow to see if it affects anything to narrow down the issue.
I don't have:
- Solder wick
- Isopropyl alcohol
- A stencil for the BGA pattern of the chip
- Little balls of solder/solder paste
- A heat gun (I have a gas soldering iron/torch which could be used for this but temperature control would all over the show and it would be super tricky to get a consistent temperature profile during the reflow!).
By the time I buy all of that I'm gonna be into this TV for the same as the cost to buy one second hand! I would love to do this BGA repair properly cos it would be really fun (although I'd exhaust my daily curse word quota on the fiddly bits! But that just makes it all the more worthwhile if you can pull it off).
Good point about not selling it if not fixing it properly. Not really fair to sell it to someone and then it craps out soon after.
-
So (to 'login's' disgust), I'll probably next try something akin to a dodgy reflow of the main BGA chip, using none of the correct tools, because I don't have them. BUT! I'll splash out and treat the board to some flux during the process.
I'm not sure if I'd try that just yet, it seems your power supply is getting overloaded which might be a different issue to the common faults people have. I would first check all the voltage regulators on the main board for input/output shorts to gnd. It could be the BGA is causing a short, try and find the datasheet then measure the resistance from its inputs to gnd as well.
Getting the power supply up and running to test 12V and 20V rails as shakalnokturn suggested is probably a good idea too. Then we could rule it out completely.
I was serious about posting on the badcaps forum too, I've used it quite a bit. Someone on there might have fixed the same model with the same symptoms. Just make sure you read their 'rules' first regarding what photos are required etc. ;)
edit: Just saw that the manual has schematics for the mainboard, that should make it a lot easier to check for shorts on the power rails of the BGA etc :D
-
Thanks, Sean. Yeah, got lots of work to do! Just trying to find time right now!
Getting the power supply up and running to test 12V and 20V rails as shakalnokturn suggested is probably a good idea too. Then we could rule it out completely.
Yeah, this makes sense. So to be clear, to do this, shakalnokturn was saying I need to:
- de-pin the power-on cable from the plug which plugs into the power supply board output connector
- connect everything else up as usual
- jump the 3.5VDC of the power supply board output connector to the power-on pin on the same connector
- then CAREFULLY test the 12v & 20v rails WITHOUT shorting them to any lower voltage rails in the process?
(I don't wanna fry anything haha! So better to double/triple check)
Provided I have good 12v and 24v rails, then...
check all the voltage regulators on the main board for input/output shorts to gnd. It could be the BGA is causing a short...measure the resistance from its inputs to gnd as well.
This is gonna test my schematic reading skills! But at least we HAVE a schematic for the main board! Will have a look through these rails etc and see what I can find. Wondering how good testing will work with these connections which may be a partial short/open circuit.
Will report back. I really wanna fix this thing! :D
EDIT: Oh, and yes, the badcaps forum seems quite good. I read some posts from there months ago when I first got this TV.
-
I'm convinced "login" must have more experience than I do with LCD TV repairs, still would be interested in the methods used, I may have things to learn from them.
I'm non learning much from the "haha, I'm just posting to say I'm going to watch you waste your time and get nowhere because in the end the mainboard is dead...".
It may well be a waste of time. (Although a failure is still a learning experience and someone learning from an experience is IMHO wiser than someone who's already the cat's whiskers.)
When I don't have a stack of known working PCB's for quick swapping, my usual order of testing is: Visual inspection, check power, thermal cam when available, check reset, check serial bus activity... I believe I'm not the only one who starts with checking power supplies.
Attached a photo of a different mainboard and PSU to illustrate the way I force the PS on. Usually I'll solder a 1kohm resistor to the standby supply then I can thread the other end to the loose wire.
If you really are convinced it's a BGA problem on the mainboard, I've had some success in confirming such cases by applying moderate pressure by thumb during a cold start (starting unplugged) and maintaining for the time a normal boot requires.. If you want to apply plenty of pressure consider sticking a wedge between PCB and LCD frame under the BGA to not distort the PCB too much.
-
Ok folks. The latest results are in. :popcorn:
Finally found some time to do some testing tonight. I went and grabbed a 1K-ohm resistor as suggested by shakalnokturn and set up a 'test rig' (see pic). By bending the legs of the resistor just so, it keeps enough spring-loaded tension on the un-soldered leg against the power-on pin to work as a primitive switch.
The needle sticking out isn't voodoo...it was so I could get a reference to the 20V without shorting things out.
So with the power-on forced using my 'switch', I got:
- 12.5VDC on the 12V pins of the output connector
- 20.7VDC on the 20V pins of the output connector
I also tested the LED driver section of the power supply board, and I got:
- 4.9VDC at the 5.7V pin on the LED driver board
- At all the Vxx pins on the LED driver board, I got between 13VDC and up to 17VDC
At the LED output connectors, I got:
- 115VDC at the grey and red cables
- Between 15VDC and 17VDC at the other cables
Seems ok to me? :-//
I started to test a few of the connectors on the mainboard. I have 3.5V, 12V and 20V all present. Didn't really know what to look for so much on the mainboard as it's not labelled like the power supply.
I took some more detailed pics of the mainboard which I'll include below this post. I've removed the heatsink and thermal sticker from the LG XD Engine chip. I had a look through the schematics. It kind of seemed like most of the places where I could find a reference for a voltage were on the underside of the mainboard when it's installed on the TV? This was based on there being a voltage listed against a component which I could physically find to use as a test point. So not sure how I'm gonna be able to test this in a powered up state?
I think I also previously posted the wrong service manual which wasn't for my TV, :-\ So I'll make a link available for that here so you can see the main board schematics (if you're reading this years from the date of this post, this link probably won't work :P ) - https://www.dropbox.com/s/idwh7ofwv4yn7vs/LG47LE7500%20Service%20Manual.pdf?dl=0
(https://www.dropbox.com/s/idwh7ofwv4yn7vs/LG47LE7500%20Service%20Manual.pdf?dl=0)
I'm a bit lost in reading these schematics, or at least knowing what I should be looking/testing for. The '3.3V normal' seemed to be mentioned a lot and that seemed important, but R1027 is one of those resistors on the underside of the board so I can't test it when it's powered up.
I'd be really grateful for any help in diagnosing the mainboard. Maybe even 'login' will see fit to share some wisdom now they seem to have been proven correct :D
-
Top of mainboard
-
Underside of mainboard
-
So with the power-on forced using my 'switch', I got:
- 12.5VDC on the 12V pins of the output connector
- 20.7VDC on the 20V pins of the output connector
I also tested the LED driver section of the power supply board, and I got:
- 4.9VDC at the 5.7V pin on the LED driver board
- At all the Vxx pins on the LED driver board, I got between 13VDC and up to 17VDC
At the LED output connectors, I got:
- 115VDC at the grey and red cables
- Between 15VDC and 17VDC at the other cables
Okay so the 3.5/12/20V rails are all ok, so that's good.
I'm pretty sure the 5.7V pin is actually 5.2V? Based on the photos in this thread:
https://www.badcaps.net/forum/showthread.php?t=52635 (https://www.badcaps.net/forum/showthread.php?t=52635)
13-17V is probably okay for the Vxx pins, but not 100% sure. Assuming you mean V3a-, V3B- etc, these are all the cathode return lines for the LED strips. It probably depends on whether the backlights are on or off.
The red and grey cables (V1+, V2+, V3+, V4+) are meant to be 136V based on the photo you posted earlier. In the bottom left corner there's jumper next to the model number which lists 136V 0.66A output. Maybe the 115V jumps to 136V once the backlights are on?
Relay RL101 would be part of the power supplies overload protection, so if it's clicking off it's because the mainboard or backlights are drawing too much current. But because you are measuring all the power supply rails while the mainboard is connected I'm starting to think it might be the backlights.
If you have the mainboard connected, but the backlights disconnected, does it still shutdown? If it doesn't shine a torch at the screen and see if you can see the logo etc.
The missing standby light is a mystery, does it come from one of the mainboard connectors? If so, which one?
-
The relay is there to enable power to the main supply through PFC.
Having the LEDs on would be better as a load test, that was the point in forcing the supply active.
The PSU seems to have a serial bus for communication with the mainboard, I assume that this is mainly related to fault detection on the LEDs but could be wrong. I also assume this should trigger an error blink with the shutdown.
Standby led comes from the mainboard microcontroller, is not just a standby pilot, it's also used for blinking error codes, so indeed failure to light at all is a bad sign for the mainboard.
On the mainboard you could check the various low supply voltages (usually between 1.2 and 3.3V) for CPU, DDR whatever... They are easy to spot on linear regulators like IC8800 others should be present on electrolytic capacitors or ferrite "beads".
Things are pointing towards the mainboard by now.
-
Having the LEDs on would be better as a load test, that was the point in forcing the supply active.
oh, but I thought it was shutting down before the backlights came on? Or are the backlights on as soon as the 115V is present?
Standby led comes from the mainboard microcontroller, is not just a standby pilot, it's also used for blinking error codes, so indeed failure to light at all is a bad sign for the mainboard.
Okay that is good to know, I agree it's a bad sign for the mainboard.
-
Hi gents. Thanks for the comments shakalnokturn and Sean. I just did some more testing based on your thoughts.
I tried powering up the TV with the backlight cables disconnected. Made no difference. The relay still clicked on and off immediately.
I then did some tests on the mainboard with the power supply forced on as per my previous post. Results were:
- 20v, 12v and 3.5v are all present at mainboard connector, AND at the 3x electrolytic capacitors immediately next to the main incoming connector. You can tell which capacitor corresponds to which voltage rail from the voltage ratings on the caps.
- I was also able to find 3.5v and 12v a little bit away from these caps doing a bit of testing, but couldn't find another 20v reference. I've attached a pic showing some areas of the board which seemed to be dead/powered off highlighted in red.
- The connector on the left side of the mainboard (P1895) seemed to have just a few millivolts registering to GND on the pins. This goes to the small board at the bottom of the TV off-centre towards the mainboard side. Not sure what this is for?
- IC2600 had essentially 0v to GND on all pins
- IC8800 had essentially 0v to GND on all pins
- The connector P8800 at the bottom is for the speakers, and had 1.4v-1.5v present on all pins to GND.
- Connector P8200 next to it which goes out to the IR receiver/power buttons/standby LED board had 3.5v on a bunch of pins, and perhaps even 12v on one (can't remember 100%, didn't take notes and tested dozens of points :palm: )
- I also tested the IR receiver/power buttons/standby LED board itself (pictures of the board below). This had 3.3v - 3.5v on at lest a couple of one pins on each connector. Both electrolytic caps had 3.3v - 3.5v. I took these pics the other day, but I haven't actually tested the LED itself or the rest of this board apart from the live testing I did tonight. Was interesting that it seems to have lots of power but not even an LED. As mentioned previously, it seems the LED is an 'all systems go' indicator for the whole TV, so there's obviously something wrong upstream which is preventing it from being switched on.
Not sure if the 20v is pretty much immediately chopped up and reduced once it goes into the mainboard, but there definitely seem to be some areas of the mainboard which are 'dead'/not turning on.
When I get a chance, I'll take another look at the schematic and see if I can figure out why it's not powering up. Might do some testing of it powered off too and see if there are any obvious shorts/open circuits etc.
Feel free to let me know any other thoughts/suggestions :D